Abstract

A big segment of the traffic signal control systems in California and United States are closed-loop systems. Because wide-scale deployment of advanced adaptive control systems may be many years away due to the associated high costs, there is a significant need to improve the effectiveness of the state-of-the-practice closed-loop systems. To address the need, this project focuses on: 1) developing an integrated micro-simulation/signal optimization tool to enhance the capability of generating efficient signal timing plans, and 2) developing a systematic approach to make closed-loop systems be more robust and traffic responsive.

An integrated simulation/signal optimization tool can generate, evaluate and fine tune signal timing plans in a cohesive manner. This report presents an approach for integrating Paramics with Synchro and TRANSYT-7F. Two sets of Paramics plug-ins are developed to facilitate the two-way data conversion between Paramics and Synchro or TRANSYT-7F. The first set of plug-ins read Paramics network data and traffic volume data and generate Synchro or TRANSYT-7F data files while the second transfer optimized signal timing data back to Paramics. These tools may assist traffic engineers in developing efficient signal timing plans for arterial traffic operations. Step-by-step tutorials are also included in the report to teach how to use the new plug-ins.

The advancement and deployment of telecommunication and ITS technologies make traffic and signal status data more readily available. These high-resolution data provide opportunities to allow closed-loop control systems to operate more adaptively and robustly to the changes in traffic demands and patterns. This report presents a systematic approach to make use of traffic and signal data to further improve the control performance of closed-loop systems. The systematic approach includes three components: timing, monitoring, and fine-tuning. For timing, two innovative models are developed to generate robust optimal signal timings that are less sensitive to fluctuations of traffic flows at the same time minimizing the mean of the delays per vehicle across all possible realizations of uncertain traffic flows. Another procedure is proposed to optimally determine time-of-day intervals for time-of-day controls based on a large set of archived traffic data. For monitoring, a prototype signal performance monitoring system is developed to report performance measures of signal control operations and help traffic operation staffs make the decision whether a retiming or fine-tuning effort is needed or not. Finally, for fine-tuning, an offset refiner is introduced to fine tune signal offsets to provide smoother progression in either one-way or two-way coordination. The offset refiner is easy to implement, and could be run periodically or together with the performance monitoring system. If the signal performance degrades, the refiner can be called to fine-tune the offsets for better progression.